Tracking on the Field

Racial Diversity and Disparity in Football Quarterbacking

Authors
Affiliations

Magdalena Bennett
Principal Investigator

The University of Texas at Austin


RESEARCH ASSISTANTS

Shishir Nambi

The University of Texas at Austin

Benny Wang

The University of Texas at Austin

Maryam Sadiq

The University of Texas at Austin

Published

January 19, 2024

Abstract
In this study, we analyze racial disparities at different levels in competitive American football: High school, College, and the National Football League (NFL). In particular, we disentangle the selection process at different stages for the most pivotal position in a team: Quarterbacks.

Introduction

The intersection of race and the quarterback position in American football has been a topic of considerable discussion and analysis over the years. Historically, the quarterback position in the National Football League (NFL) was predominantly occupied by white athletes, reflecting broader societal attitudes and stereotypes about leadership and intellectual roles in sports. However, the landscape has been changing gradually.

Recent years have witnessed a significant shift in this trend. As of 2023, the NFL has seen a notable increase in the number of quarterbacks of African American and other racial backgrounds leading teams, challenging long-standing biases. For instance, at the start of the 2023 season, approximately 44% of starting quarterbacks in the NFL were African American, a record for the league and a marked increase from just 25% two decades earlier. This change not only reflects a shift in the racial dynamics within the sport but also indicates a broader societal progression towards inclusivity and recognition of diverse talents.

This study aims to delve deeper into the statistical trends of potential racial discrimination in the quarterback position, analyzing selection at different stages of a player’s career: from high school to college, and, potentially, from college to pro-ball. In particular, we will analyze shifts and selections at these two stages, shown in the following plot:

Figure 1. Selection process for Quarterbacks at different stages by race

Data

Sources

To analyze the progression of racial disparity in the quarterback position, we merge several data sources:

  1. Quarterback Stats for College Football (1980 - 2019): This dataset contains data from sports-reference.com about quarterbacks that play college football, including some key statistics such as number of games, passing yards, touchdowns, etc. Statistics from 2017 to 2019 were downloaded directly from the website.

  2. College Football Total QBR (2007-2019): Dataset from espn.com, including 135 QBs per year, their ranking, and other statistics.

  3. ESPN Top Quarterbacks (2006 - 2017): Dataset from espn.com, which includes the top 100 recruits by position (QBs) each year. This dataset includes information about their college selection (if any), weight, height, high school, grade, and ranking.

  4. College Football Stats (2006-2019): Dataset from espn.com, with information about passing, rushing, and receiving leaderboards for each year. It also includes specifific statistics for each board, such as passing/rushing yards, touchdowns, and interceptions, among others. These datasets are also used to identify the position of each player in college. As a reference, the passing leaderboard includes around 500 players per year (mostly QBs), while the rushing and receiving leaderboard include over 1200 players each.

  5. College Football Team Ranking (2005-2019): Information at the college team level for each year, including the Football Power Index metric from ESPN, as well as the record of wins and losses for the season.

  6. Player information for College Football (1980 - 2022): This is a dataset gathered by a Reddit user (see here), including information about their position, statistics, school and town of origin, and some sparse NFL statistics as well.

  7. Draft picks (1980-2022): Dataset with all draft picks by year and position.

  8. Maxpreps data for top high school quarterbacks (2012 - 2022): Information about top 30 high school quarterbacks each year (according to maxprep ranking), including school and statistics. This has been complemented by a manual search of the player’s current status and potential position change.

  9. Perceived race classification (2006 - 2019): This is a variable manually gathered for QBs listed in the first and third data source. This is a perceived race classification based on images of the players, with three distinct categories: White, Black, and Other. Even though this variable does not capture the actual race of the player, it will be a proxy for race perception and how this aspect affects their selection at different stages.

We are currently working on harmonizing all these datasets, and adding high school demographic information for each player, in addition to extending the perceived race classification for additional years.

Data Description

High School Level

At the high school level, we are working with the top Quarterback recruits according to ESPN (100 to 150 each year), to analyze their potential path into college football. In total, we have 1386 unique high school QBs between 2006 and 2017. The following table shows the characteristics that are available in the data.

Table 1. Variable description and distribution for Top HS QBs (Source: ESPN)

College Level

In total, we have 1960 individual college quarterbacks between the years of 2007 and 2019, that are distributed throughout the US as shown in the following map.

Figure 2. College location for 2010-2016 data

We also have the following characteristics for each of the players, as described in the table below:

Table 2. Variables description and distribution for College QBs by category (Source: sports-reference.com and ESPN)

Preliminary Results

High School Level

As a first approach, we will analyze the top quarterback recruits between 2006 and 2017, and follow their progression into college football. By focusing on the top recruit prospects (as defined by ESPN), we are able to analyze the top contenders for QB positions each year. In particular, we will be focused on two specific questions:

  1. Is there a disparity in recruitment by race?
  2. If recruited, is there a shift in position that is different by race?

The first question addresses the issue of potential discrimination on the extensive margin, and the second one tackles the issue of potential discrimination in terms of the specific QB position.

In our data, 63% of top recruits end up accepting an offer from a college to play football, and there is no significant difference by race. In a future iteration, we will add characteristics at the college level to analyze whether there are systematic differences in the quality of teams that students are recruited to.

Regarding shifts or changes in position, however, the story is very different. While the majority of top high school QBs are recruited to attend college to play football, only 85% of those recruited end up in the quarterback position in college.

The following table shows the comparison between white and non-white top recruits in terms of some of their characteristics. As show in this table, even though both groups are comparable in terms of grade and physical attributes, ranked QBs of color have a much higher ranking than white quarterbacks. There is also a significant difference in how they are classified since 2013 onward: Pocket passers or dual threats. Additionally, there is a significant difference in the probability of being recruited as a QB in college, but not in being recruited for college football.

Table 3. Comparison of characteristics between white vs non-white Top QB Recruits (2006-2017) (Source: ESPN)
White (N=927)
Non-white (N=459)
Mean Std. Dev. Mean Std. Dev. Diff. in Means p
Ranking 46.02 28.04 39.00 24.62 -7.02 <0.01
Grade 76.05 4.57 76.14 4.44 0.09 0.72
Height 6.22 0.13 6.17 0.15 -0.06 <0.01
Weight 198.36 14.83 194.80 16.83 -3.56 <0.01
Dual Threat List 0.11 0.31 0.45 0.50 0.34 <0.01
Pocket Passer List 0.35 0.48 0.11 0.31 -0.24 <0.01
Not Classified (pre-2013) 0.54 0.50 0.44 0.50 -0.10 <0.01
College Recruit 0.64 0.48 0.62 0.49 -0.02 0.50
Recruited for College as QB 0.55 0.50 0.42 0.49 -0.13 <0.01
Recruited for College not as QB 0.09 0.28 0.20 0.40 0.11 <0.01

We ran four different models to analyze the difference in probability of being recruited as a college QB by race. Results show that high school quarterbacks of color are, on average, 14 percentage points less likely to be recruited as a QB for college football, even after controlling for performance and physical characteristics. We also included their play style classification (dual threat or pocket passers) for QBs from 2013 onward, and the difference in probability still remains. These players will then move on to other positions during their college years, which increases the disparities in the next selection stage, the NFL draft.

Table 4. Linear Probability Models for being recruited as a QB in College as a Top 100 HS QB, based on race and performance
College QB College QB College QB College QB
NonWhiteQB −0.136*** −0.139*** −0.140*** −0.129**
(0.028) (0.027) (0.028) (0.046)
Grade 0.031*** 0.032*** 0.028***
(0.003) (0.003) (0.003)
Height 0.095 0.011
(0.106) (0.154)
Weight −0.002* −0.004**
(0.001) (0.001)
Dual Threat −0.086+
(0.047)
N 1386 1386 1386 686
Avg. Outcome (Control) 0.553 0.553 0.553 0.579
Year FE Yes Yes Yes Yes
Years 06-17 06-17 06-17 13-17
R2 Adj. 0.521 0.559 0.560 0.575
+ p < 0.1, * p < 0.05, ** p < 0.01, *** p < 0.001

In order to analyze the heterogeneity across years, we ran the previous regression (column 3) separately for different years. The following figure shows the estimated coefficients for NonWhiteQB for three specific time periods: 2006 to 2010, 2011 to 2014, and 2015 to 2017. As a reference, 2011 was the year Cam Newton was drafted into the NFL.

Figure 3. Difference in probability of QB college recruitment for non-white HS QBs

Given that white and non-white QBs could also have different styles of play, where one could be better rewarded because of their style, we re-weight non-white quarterback observations to match the characteristics of white QBs in terms of overall scouts’ grade, style of play (e.g. pocket passers or dual threats), weight, height, and year of graduation.

We follow Chattopadhyay, Hase, and Zubizarreta (2020) for building stable weights that approximately balance the empirical distribution of the aforementioned covariates with a tolerance of 0.01 SD, using the distribution of performance for white high school QBs as the target. The following table shows the balance between both groups after weighting.

Table 5. Average charactersitics between white and non-white HS QBs after Stable Balance Weighting
White HS QB Non-white HS QB Target Diff in Means (SD)
Year 2011.66 2011.65 2011.66 0.002
Grade 76.05 76.09 76.05 0.010
Weight (lbs) 198.37 198.21 198.37 0.010
Height (ft) 6.22 6.22 6.22 0.010
Dt 0.11 0.11 0.11 0.010
Pocket passer list 0.35 0.35 0.35 0.010
Not classified 0.54 0.54 0.54 0.004


Comparing the probability of being recruited to play in college as a QB for both groups after weighting, we find that there is a 0.106 (SE = 0.017; p-value = 0) decrease in the average probability of being recruited for high school QBs of color compared to white QBs, which is smaller than our previous result, but still a large difference.

College Level

After analyzing the leaks in the pipeline between high school and college, we will now move on to analyzing players during their college years. In this case, we will compare the performance in college football between white quarterbacks and quarterbacks of color. From the following table, we can see that overall, non-white QBs perform similarly than white QBs. One key difference between both groups, though, is that non-white QBs have significantly more running yards in comparison with passing attempts and passing yards, which could also demonstrate a difference in playing style between both. In terms of touch downs, however, both groups have similar performance.

Table 6. Comparison of characteristics between white vs non-white College QB (2006-2019) (Source: sports-reference.com and ESPN)
White (N=2887)
Non-white (N=1093)
Mean Std. Dev. Mean Std. Dev. Diff. in Means p
Rank 63.15 34.85 59.70 35.87 -3.45 0.09
QBR 56.00 16.31 58.01 16.74 2.01 0.04
Raw Rank (ESPN) 55.21 15.08 56.59 15.14 1.38 0.11
Plays 456.55 116.49 456.90 120.80 0.36 0.96
Points contributed 13.24 30.44 15.89 31.36 2.66 0.14
Pass 36.15 24.48 31.85 22.75 -4.30 <0.01
Run 4.55 10.19 10.83 13.35 6.28 <0.01
Sack -13.05 5.61 -12.77 5.33 0.27 0.38
Pen 1.77 2.15 1.50 2.20 -0.26 0.04
Drafted 0.15 0.36 0.15 0.36 0.00 0.91
Pick (Draft) 123.73 81.68 108.78 83.14 -14.95 0.05
Round (Draft) 4.10 2.13 3.65 2.22 -0.45 0.03


Figure 4. SD Difference in QB charactersitics by race (non-white vs white)


Looking at the distribution of performance, we can observe that QBs of color follow a bi-modal distribution, but have an average performance in terms of QB rating that is similar (or even slightly higher) than white quarterbacks.

Figure 5. Distribution of QB rating by race (Source: ESPN)


We observe a similar trend when analyzing points contributed as a function of QB rating. While there is a clear positive association between both, there does not appear to be a significant difference in terms of distribution by race.

Figure 6. Number of points contributed vs QB Rating (Source: ESPN)


In terms of their professional prospects, though, measured as being drafted by an NFL team as a quarterback, results differ between both groups. The following models shows the difference in probability of being drafted (NonWhiteQB) for non-white QBs compared to white QBs. Even after controlling for year and some performance metrics, such as QB rating and passing touch downs, the differences persist, though these estimates are not statistically significant at conventional levels for all specifications:

Table 7. Linear Probability Models for being drafted as a QB for College QBs, based on race and performance
Drafted Drafted Drafted Drafted
NonWhiteQB 0.001 −0.061+ −0.059+ −0.047
(0.021) (0.036) (0.035) (0.035)
QBR 0.009*** 0.008*** 0.006***
(0.001) (0.001) (0.001)
Plays 0.000*** 0.000
(0.000) (0.000)
Pass 0.002+
(0.001)
N 3980 1488 1488 1488
Avg. Outcome (Control) 0.15 0.286 0.286 0.286
Year FE Yes Yes Yes Yes
R2 Adj. 0.149 0.353 0.360 0.362
+ p < 0.1, * p < 0.05, ** p < 0.01, *** p < 0.001


On average, a white college QB has a 29% probability of being drafted in our dataset1, which means that a negative effect between 4.7 and 6.1 percentage points on being drafted represents a decrease in probability between 16% and 21% associated to being a QB of color.

In order to match the play style between white QBs and QBs of color, we perform the same re-weighting exercised as before, matching the style of play of white QBs in terms of passing yards/attempts, rushing yards/attempts, interceptions, touch downs, games played, and other aggregated performance characteristics such as QB Rating, raw rating, and points contributed.


The following table shows the balance between both groups after weighting.

Table 8. Average charactersitics between white and non-white QBs after Stable Balance Weighting
White QB Non-white QB Target Diff in Means (SD)
Plays 456.55 457.76 456.55 0.01
Completions 210.80 210.07 210.80 0.01
Pass 36.15 35.92 36.15 0.01
Yards (total) 2550.88 2542.27 2550.88 0.01
Yards (avg) 7.30 7.31 7.30 0.01
Interceptions 9.31 9.28 9.31 0.01
Run 4.55 4.69 4.55 0.01
QBR 56.00 56.17 56.00 0.01
Rating (passing) 133.02 133.24 133.02 0.01
Points contributed 13.24 13.24 13.24 0.00
Raw Rank (ESPN) 55.21 55.06 55.21 0.01
TDs 18.21 18.30 18.21 0.01


After weighting for style of play, we find that there is a 0.067 (SE = 0.015; p-value = 0) decrease in the average probability of being drafted for QBs of color compared to white QBs. This difference is similar in magnitude to our previous results, but in this case is statistically significant. These findings indicate that even though there could be other potential confounders in play, playing style as captured by these main metrics are not the ones driving this significant difference in drafting probability.

Footnotes

  1. This result is for the top 100 ranked QBs in college↩︎